Monolayer Growth and Exchange Kinetics for Alkylamines on the High

Langmuir 2000, 16, 2169-2176

2169

Monolayer Growth and Exchange Kinetics for Alkylamines on the High-Temperature Superconductor YBa2Cu3O7-δ Feng Xu, Jin Zhu, and Chad A. Mirkin* Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208-3113 Received July 12, 1999. In Final Form: November 4, 1999 Using redox-active alkylamine adsorbates, we have studied monolayer growth and adsorbate-adsorbate exchange kinetics on YBa2Cu3O7-δ. Herein we show that the formation kinetics for monolayers of FcC(O)NH(CH2)4NH2 [Fc ) (η5-C5H4)Fe(η5-C5H5)] do not follow typical Langmuir adsorption isotherm behavior. Specifically, a parabolic tailing in the isotherms has been attributed to the diffusion of adsorbates through the porous material. The displacement of surface-adsorbed FcC(O)NH(CH2)4NH2 by another redox-active alkylamine, Fc(CH2)6NH2, has been studied and determined to be a reversible and dynamic process. Concentration dependence studies suggest that the exchange takes place via an associative process involving surface Cu(II) sites. The lability of the surface-adsorbed alkylamine “ligands” and the proposed exchange mechanism are consistent with the solution coordination chemistry of Cu(II).

Introduction Surface coordination chemistry and modification studies have led to the development of methods for tailoring the surface and interfacial properties of many important materials, some of which include Au,1 Ag,2 Pt,3 CdS,4 CdSe,4 GaAs,5 Si,6 and TiO2.7 Chemically modified surfaces have provided important insight in the areas of interfacial electron transfer,8 Raman spectroscopy,9 tribology,10 and, * To whom correspondence should be addressed. E-mail: [email protected]. Phone: (847) 491 2907. Fax: (847) 467 5123. (1) (a) Ulman, A. An Introduction to Ultrathin Organic Films: From Langmuir-Blodgett to Self-assembly; Academic: Boston, 1991. (b) Bain, C. D.; Whitesides, G. M. Angew. Chem., Int. Ed. Engl. 1989, 28, 506. (c) Dubois, L. H.; Nuzzo, R. G. Annu. Rev. Phys. Chem. 1992, 43, 437. (d) Ulman, A. Chem. Rev. 1996, 96, 1533. (2) (a) Laibinis, P. E.; Fox, M. A.; Folkers, J. P.; Whitesides, G. M. Langmuir 1991, 7, 3167. (b) Sellers, H.; Ulman, A.; Shnidman, Y.; Eilers, J. E. J. Am. Chem. Soc. 1993, 115, 9389. (c) Ye, Q.; Fang, J. X.; Sun, L J. Phys. Chem. B. 1997, 101, 8821. (d) Walczak, M. M.; Chung, C.; Stole, S. M.; Widrig, C. A.; Porter, M. D. J. Am. Chem. Soc. 1991, 113, 2370. (3) (a) Lee, T. R.; Laibinis, P. E.; Folkers, J. P.; Whitesides, G. M. Pure Appl. Chem. 1991, 63, 821. (b) Lane, R. F.; Hubbard, A. T. J. Phys. Chem. 1973, 77, 1401. (c) Hickman, J. J.; Zou, C.; Ofer, D.; Harvey, P. D.; Wrighton, M. S.; Laibinis, P. E.; Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1989, 111, 7271. (d) Stern, D. A.; Laguren-Davidson, L.; Frank, D. G.; Gui, J. Y.; Lin, C. H.; Lu, F.; Salaita, G. N.; Walton, N.; Zapien, D. C.; Hubbard, A. T. J. Am. Chem. Soc. 1989, 111, 877. (4) Natan, M. J.; Thackeray, J. W.; Wrighton, M. S. J. Phys. Chem. 1986, 90, 4089. (5) Sheen, C. W.; Shi, J.-X.; Martensson, J.; Parikh, A. N.; Allara, D. L. J. Am. Chem. Soc. 1992, 114, 1514. (6) (a) Linford, M. R.; Chidsey, C. E. D. J. Am. Chem. Soc. 1993, 115, 5, 12631. (b) Keller, H.; Schrepp, W.; Fuchs, H. Thin Solid Films 1992, 210/211, 799. (7) (a) Meyer, T. J.; Meyer, G. J.; Pfennig, B. W.; Schoonover, J. R.; Timpson, C. J.; Wall, J. F.; Obusch, C.; Chen, X. H.; Peek, B. M.; Wall, C. G.; Ou, W.; Erickson, B. W.; Bignozzi, C. A. Inorg. Chem. 1994, 33, 3952. (b) Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 45. (b) Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 52. (c) Ogawa, H.; Chihera, T.; Taya, K. J. Am. Chem. Soc. 1985, 107, 1365. (8) (a) Finklea, H. O. In Electroanalytical Chemistry; Bard, A. J., Ed.; Marcel Dekker: New York, 1995; Vol. 19, Chapter II. (b) Campbell, D. J.; Herr, B. R.; Hulteen, J. C.; Van Duyne, R. P.; Mirkin, C. A. J. Am. Chem. Soc. 1996, 118, 10211. (c) Weber, K.; Hockett, L. A.; Creager, S. E. J. Phys. Chem. B 1997, 101, 8286. (d) Chidsey, C. E. D. Science 1991, 251, 919. (9) (a) Freeman, R. G.; Grabar, K. C.; Allison, K. J.; Bright, R. M.; Davis, J. A.; Guthrie, A. P.; Hommer, M. B.; Jackson, M. A.; Smith, P. C.; Walter, D. G.; Natan, M. J. Science 1995, 267, 162. (b) Zhu, J.; Xu, F.; Schofer, S. J.; Mirkin, C. A. J. Am. Chem. Soc. 1997, 119, 235.

chemical force microscopy.11 In addition, they have resulted in the development of many useful materials and technologies including sensors,12,13 catalysts,14 nonlinear optical materials,15 lithographic resists,16 nanolithography methods,17 and optoelectronic materials.13 Recently, we developed a methodology for modifying cuprate high-temperature superconductors (HTSCs) with a variety of organic adsorbates.18 Although these studies showed that alkanethiols, selenols, and amines all adsorb onto the surfaces of cuprate-based materials, the alkylamine reagent has emerged as the most efficient surface (10) Carpick, R. W.; Salmeron, M. Chem. Rev. 1997, 97, 1163. (11) (a) Frisbie, C. D.; Rozsnyai, L. F.; Noy, A.; Wrighton, M. S.; Lieber, C. M. Science 1994, 265, 5181. (b) Noy, A.; Vezenov, D. V.; Lieber, C. M. Annu. Rev. Mater. Sci. 1997, 27, 381. (c) McKendry, R.; Theoclitou, M. E.; Rayment T.; Abell, C. Nature 1998, 391, 566. (12) (a) Mirkin, C. A.; Ratner, M. A. Annu. Rev. Phys. Chem. 1992, 43, 719. (b) Mirkin, C. A.; Valentine, J. R.; Ofer, D.; Hickman, J. J.; Wrighton, M. S. In Biosensors and Chemical Sensors; Edelman, P. G., Wang J., Eds.; ACS Symposium Series 487; American Chemical Society: Washington, DC, 1992; Chapter 17. (c) Hickman, J. J.; Ofer, D.; Laibinis, P. E.; Whitesides, G. M.; Wrighton, M. S. Science 1991, 252, 688. (d) Wells, M.; Crooks, R. M. J. Am. Chem. Soc. 1996, 118, 3988. (e) Crooks, R. M.; Ricco, A. J. Acc. Chem. Res. 1998, 31, 219. (13) (a) Storhoff, J. J.; Mucic, R. C.; Mirkin, C. A.; Letsinger, R. L. J. Am. Chem. Soc. 1998, 120, 12674. (b) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382, 607-609. (c) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078-1080. (d) Storhoff, J. J.; Mucic, R. C.; Mirkin, C. A. J. Clust. Sci. 1997, 8, 179-216. (e) Hostetler, M. J.; Murray, R. W. Curr. Opin. Colloid Interface Sci. 1997, 2, 42. (14) (a) Dalmia, A.; Liu, C. C.; Savinell, R. F. J. Electroanal. Chem. 1997, 430,205. (b) Willner, I.; Lapidot, N.; Riklin, A.; Kasher, R.; Zanavy, E.; Katz, E. J. Am. Chem. Soc. 1994, 116, 1428. (c) Hutchison, J. E.; Postlethwaite, T. A.; Murray, R. W. Langmuir 1993, 9, 3277. (15) Marks, T. J.; Ratner, M. A. Angew. Chem., Int. Ed. Engl. 1995, 34, 155. (16) (a) Komeda, T.; Namba, K.; Nishioka, Y. J. Vac. Sci. Technol. A 1998, 16, 1680. (b) Tiberio, R. C.; Craighead, H. G.; Lercel, M.; Lau, T.; Sheen, C. W.; Allara, D. L. Appl. Phys. Lett. 1993, 62, 476. (c) Schoer, J. K.; Crooks, R. M. Langmuir 1997, 13, 2323. (d) Behm, J. M.; Lykke, K. R.; Pellin, M. J.; Hemminger, J. C. Langmuir 1996, 12, 2121. (e) Xu, S.; Liu, G. Langmuir 1997, 13, 127. (17) (a) Piner, R. D.; Zhu, J.; Xu, F.; Hong, S.; Mirkin, C. A. Science 1999, 283, 661. (b) Xia, Y.; Whitesides, G. M. Angew. Chem., Int. Ed. Engl. 1998, 37, 550 and references therein. (18) (a) Xu, F.; Chen, K.; Piner, R. D.; Mirkin, C. A.; Ritchie, J. E.; McDevitt, J. E.; Cannon, M. O.; Kanis, D. Langmuir 1998, 14, 6505. (b) Chen, K.; Mirkin, C. A.; Lo, R.; Zhao, J.; McDevitt, J. T. J. Am. Chem. Soc. 1995, 117, 6374. (c) For a review on the surface modification chemistry of cuprate HTSCs, see: Mirkin, C. A.; Xu, F.; Zhu, J. Adv. Mater. 1997, 9 (2), 167.

10.1021/la990910y CCC: $19.00 © 2000 American Chemical Society Published on Web 01/04/2000

2170

Langmuir, Vol. 16, No. 5, 2000 Scheme 1

Xu et al.

monolayer formation, the mechanistic nature of the adsorbate exchange process, and the factors that contribute to the anomalously large surface coverages obtained from this type of surface modification chemistry.18 Experimental Section

binder and is the adsorbate that leads to the most robust monolayer structures on YBa2Cu3O7-δ, the most extensively studied and commercially most promising HTSC material. The reaction between YBa2Cu3O7-δ and an alkylamine is not just simple coordination chemistry. Indeed, other work has shown that the first step in the process involves a sacrificial redox reaction between the alkylamine and superconductor19a (Scheme 1). The second step involves another equivalent of amine binding to the resulting Cu2+ sites at the reduced superconductor surface.18,19a This has been confirmed by secondary ion mass spectrometery,19a X-ray photoelectron spectrometry,19a and surface-enhanced Raman spectroscopy.19b Nevertheless, the oxidative step in this proposed reaction pathway has recently been challenged by McDevitt et al.,19c but in that work the authors admittedly used corroded cuprate materials where the surface already was reduced, a fact that is clearly evident in a comparison of the XPS spectra of their starting materials with ours.19a,c A simple experiment involving the soaking of an amine in the presence of freshly prepared high surface area YBa2Cu3O7-δ powder under anaerobic conditions unambiguously shows that the HTSC is capable of oxidizing the amine under the conditions used to modify its surface.19a Other than the chemistry described above, very little is known about the dynamics associated with the monolayer formation process. The rates of the reaction under well-defined conditions, the mobility of the adsorbates on the surface of the superconductor surface, and the susceptibility of the surface bound adsorbates to exchange with solution adsorbates have not been evaluated. In contrast, much is known about the analogous processes involving alkanethiol adsorption,20 migration,21 and exchange on Au substrates,22 and this information has been used to better understand the monolayer formation process and the environmental conditions that govern monolayer stability. In this paper, we explore the monolayer growth and exchange rates for alkylamines on YBa2Cu3O7-δ. In addition, we comment on the factors that influence (19) (a) Zhu, J.; Mirkin, C. A.; Braun, R. M.; Winograd, N. J. Am. Chem. Soc. 1998, 120, 5126. (b) Zhu, J.; Xu, F.; Schofer, S. J.; Mirkin, C. A. J. Am. Chem. Soc. 1997, 119, 235. (c) Ritchie, J. E.; Murray, W. R.; Kershan, K.; Diaz, V.; Tran, L.; McDevitt, J. T. J. Am. Chem. Soc. 1999, 121, 7447.

General Methods. All manipulations were performed under air- and moisture-free conditions using standard Schlenk techniques or in an inert atmosphere glovebox, unless otherwise noted. Tetrabutylammonium hexafluorophosphate (n-Bu4NPF6), obtained from Aldrich Chemical Co., was recrystallized three times from ethanol and stored in a nitrogen atmosphere glovebox prior to use. Acetonitrile (CH3CN) was dried by refluxing over calcium hydride, while diethyl ether (Et2O) was dried and distilled over sodium-benzophenone. YBa2Cu3O7-δ superconductor powder (99.999%, particle size 0.2 µm) was purchased from Strem Chemicals, Inc. NMR spectra were recorded on a Varian Gemini300 FT NMR spectrometer. Electron impact (EI) mass spectra were obtained using a Fisions VG 70-250 SE mass spectrometer. Infrared absorption spectra were recorded on a Nicolet 520 Fourier transform infrared spectrometer with a liquid nitrogen cooled MCT (HgCdTe) detector. 6-Ferrocenylhexyl bromide was synthesized via modifications of a literature method.23 The preparation of FcC(O)NH(CH2)4NH2 (compound 1) has been reported previously.18a 1H NMR and mass spectra of these compounds match those reported in the literature. Synthesis of 6-Aminohexylferrocene (Compound 2). 6-Ferrocenylhexyl bromide (1.25 g, 3.6 mmol) was dissolved in 50 mL of N,N′-dimethylformamide (DMF) in a 100 mL roundbottom flask. NaN3 (0.26 g, 4.0 mmol) was added to the solution via a powder addition funnel, and the mixture was stirred at room temperature for 12 h. The reaction mixture was subsequently quenched with 100 mL of water. The solution was extracted with ethyl acetate (3 × 100 mL) and washed with brine (3 × 100 mL). After the organic extract was dried over magnesium sulfate, the solvent was removed by rotary evaporation, yielding 6-azidohexylferrocene (1.0 g, 3.2 mmol, yield ) 89%). 1H NMR (C6D6): δ 4.02 (s, 5H, η5-C5H5), 3.97 (m, 4H, η5-C5H4), 2.73 (t, 2H, CH2N3), 2.19 (t, 2H, η5-C5H4CH2), 1.36 (m, 2H, CH2CH2N3), 1.20 (m, 2H, η5-C5H4CH2CH2), 1.08 (m, 4H, CH2CH2CH2CH2N3). HRMS (EI) (M+) calcd for C16H21N3Fe: 311.108 m/z. Found: 311.109 m/z. 6-Azidohexylferrocene (1.0 g, 3.2 mmol) was placed in a 100 mL Schlenk flask and dissolved in 50 mL of dry Et2O. LiAlH4 (0.18 g, 4.8 mmol) was added to a separate 100 mL Schlenk flask and suspended in 30 mL of Et2O. The solution containing 6-azidohexylferrocene was transferred dropwise to the LiAlH4 suspension via a cannula. The mixture was stirred under nitrogen for 2 h and then quenched with 20 mL of 1 mM aqueous NaOH. Fifty milliliters of Et2O was added to extract the ferrocenylcontaining product from the aqueous suspension. The orange extract was dried over magnesium sulfate. Rotary evaporation of the solvent yielded the desired product as a yellow oil (0.82 g, 2.88 mmol, yield ) 90%). 1H NMR (C6D6): δ 4.01 (s, 5H, η5(20) (a) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321. (b) Biekuyck, H. A.; Bain, C. D.; Whitesides, G. M. Langmuir 1994, 10, 1825. (c) Schneider, T. W.; Buttry, D. A. J. Am. Chem. Soc. 1993, 115, 12391. (d) Forouzan, F.; Bard, A. J.; Mirkin, M. V. Isr. J. Chem. 1997, 37, 155. (e) Kim, Y.-T.; McCarley, R. L.; Bard, A. J. Langmuir 1993, 9, 1991. (f) Schessler, H. M.; Karpovich, D. S.; Blanchard, G. J. J. Am. Chem. Soc. 1996, 118, 9645. (g) Pan, W.; Durning, C. J.; Turro, N. J. Langmuir 1996, 12, 4469. (h) Schlenoff, J. B.; Li, M.; Ly, H. J. Am. Chem. Soc. 1995, 117, 12528. (i) Kajikawa, K.; Hara, M.; Sasabe, H.; Knoll, W. Jpn. J. Appl. Phys. 2 1997, 36, L1116. (j) Thomas, R. C.; Sun, L.; Crooks, R. M. Langmuir 1991, 7, 620. (k) Bensebaa, F.; Voicu, R.; Huron, L.; Ellis, T. H. Langmuir 1997, 13, 5335. (21) (a) Kondoh, H.; Kodama, C.; Nozoye, H.; J. Phys. Chem. B. 1998, 102, 2310. (b) Cavalleri, O.; Hirstein, A.; Bucher, J. P.; Kern, K. Thin Solid Films 1996, 285, 392. (c) Sondaghuetorst, J. A. M.; Schonenberger, C.; Fokkink, L. G. J. J. Phys. Chem. 1994, 98, 6826. (22) (a) Collard, D. M.; Fox, M. A. Langmuir 1991, 7, 1192. (b) Chidsey, C. E. D.; Carolyn, R. B.; Putvinski, T. M.; Mujsce, A. M. J. Am. Chem. Soc. 1990, 112, 4301. (c) Rowe, G. K.; Creager, S. E. Langmuir 1994, 10, 1186. (d) Hickman, J. J.; Ofer, D.; Zou, C. F.; Wrighton, M. S.; Laibinis, P. E.; Whitesides, G. M. J. Am. Chem. Soc. 1991, 113, 1128. (23) Creager, S. E.; Rowe, G. K. J. Electroanal. Chem. 1994, 370, 203.

Growth and Exchange Kinetics for Alkylamines C5H5), 3.97 (m, 4H, η5-C5H4), 2.52 (t, 2H, CH2NH2), 2.25 (t, 2H, η5-C5H4CH2), 1.48 (m, 2H, CH2CH2NH2), 1.26 (m, 6H, η5-C5H4CH2CH2 CH2CH2), 0.65 (s, br, 2H, NH2). HRMS (EI) (M+) calcd for C16H23NFe: 285.118 m/z. Found: 285.117 m/z. Preparation of YBa2Cu3O7-δ Electrodes. YBa2Cu3O7-δ powder was pressed into 13 mm diameter pellets (applied pressure ∼4000 psi, ∼1 mm thick), which were subsequently annealed at ∼450 °C overnight under an oxygen flow (pressure ) 1 atm), followed by slow cooling (18 °C/h), under oxygen flow, to room temperature. All of the resulting pellet samples displayed a superconducting transition temperature of ∼92 K. XRD studies showed that the pellets prepared under these conditions are single-phase, orthorhombic YBa2Cu3O7-δ. Pellets were then fabricated into epoxy-encapsulated electrodes via literature methods.24 Procedure for Modifying YBa2Cu3O7-δ Electrodes with Alkylamines. In a typical experiment, an epoxy-encapsulated YBa2Cu3O7-δ ceramic electrode was cut with a razor blade to expose a fresh surface on the tip of the electrode. The exposed surface was then sanded and flattened with 600 grit sand paper, followed by sonication in dry CH3CN for 1 min. The electrode was shaved with a new razor blade according to literature methods.18b The electrode was then immediately transferred into a glovebox filled with nitrogen. The freshly prepared YBa2Cu3O7-δ ceramic electrode was immersed into a dry CH3CN solution of alkylamine for a set period of time, which depended upon the particular experiment (vide infra). The modified electrode was removed from the soaking solution and rinsed throughly with copious amounts of dry CH3CN prior to performing electrochemical measurements. After each electrochemical measurement, the electrode was rinsed with fresh CH3CN to remove electrolyte before immersing it into a new CH3CN solution of alkylamine adsorbate reagent for further studies. Electrochemical Measurements. Cyclic voltammetry was performed with a Pine AFRDE4 bipotentiostat with a Linseis X-Y 1400 recorder. All electrochemical measurements were performed with a conventional three-electrode cell in a glovebox. Each cell consisted of a modified or unmodified YBa2Cu3O7-δ ceramic working electrode, a Pt gauze counter electrode, and a nonaqueous Ag/AgNO3 (0.01 M AgNO3/CH3CN) reference electrode. The electrolyte solution was 0.1 M n-Bu4NPF6/CH3CN for all electrochemical experiments. Scanning Electron Microscopy. Morphologies of ceramic YBa2Cu3O7-δ electrodes were examined by a Hitachi S570 scanning electron microscope with an accelerating voltage of 25 keV. The working distance beteeen the electron gun and sample surface was in the 25-30 mm range, and samples were connected to a metal sample holder with a conducting graphite tape.

Results and Discussion To probe monolayer growth and exchange kinetics, two ferrocenyl alkylamine adsorbates, FcC(O)NH(CH2)4NH2 (1) and Fc(CH2)6NH2 (2), were synthesized. Ferrocenyl-

containing adsorbates have been extensively studied in the preparation and characterization of SAMs on various inorganic substrates.2,25 Indeed, the surface coverage (Γ) (24) Riley, D. R.; McDevitt, J. T. J. Electroanal. Chem. 1990, 295, 373. (25) (a) Hickman, J. J.; Laibinis, P. E.; Auerbach, D. I.; Zou, C.; Gardner, T. J.; Whitesides, G. M.; Wrighton, M. S. Langmuir 1992, 8, 357. (b) Shimazu, K.; Sato, Y.; Yagi, I.; Uosaki, K. Bull. Chem. Soc. Jpn. 1994, 67, 863. (c) Gardner, T. J.; Frisbie, C. D.; Wrighton, M. S. J. Am. Chem. Soc. 1995, 117, 6927. (d) Caldwell, W. B.; Campbell, D. J.; Chen, K.; Herr, B. R.; Mirkin, C. A.; Malik, A.; Durbin, M. K.; Huang, K. G.; Dutta, P. J. Am. Chem. Soc. 1995, 117, 6071.

Langmuir, Vol. 16, No. 5, 2000 2171 Scheme 2

of a ferrocenyl-containing adsorbate can be determined via eq 1,

Γ ) Q/nFA

(1)

where n is the number of electrons involved in the electrontransfer process (n ) 1 for ferrocene/ferrocenium redox couple), F is the Faraday constant, Q is the charge passed across the electrode/monolayer interface, and A is the geometric surface area of the electrode (vide infra). The charge Q can be determined by integrating the current associated with the ferrocenyl/ferrocenylium redox reaction when probed by cyclic voltammetry. One key assumption in this analysis is that all of the surfaceadsorbed ferrocenyl alkylamines are electrochemically accessible. Although not safe for all adsorbates,8b,26 this assumption has proven reasonable for monolayer studies on gold involving ferrocenyl alkyl adsorbates with less than 10 methylene units.8a The monolayer formation rate can be determined by measuring the change in surface coverage for ferrocenyl-containing adsorbate as a function of adsorption time (case 1 in Scheme 2). For compound 1, a strongly electron-withdrawing amido functional group has been attached to one of the cyclopentadienyl rings of the ferrocenyl moiety, which makes the ferrocenyl group more electron deficient and thus more difficult to oxidize than ferrocene. For compound 2, a slightly electrondonating methylene group is bound to one cyclopentadienyl ring, making it slightly easier to oxidize than ferrocene. Due to the different electronic influences of these two substituents, the redox potentials of 1 (E1/2 - 300 mV vs Ag/AgNO3) and 2 (E1/2 - 80 mV vs Ag/AgNO3) are separated by 220 mV in 0.1 M n-Bu4NPF6/CH3CN (Figure 1). This difference in redox potential allows one to study adsorbate exchange reactions on YBa2Cu3O7-δ via cyclic voltammetry (case 2 in Scheme 2). In general, one can examine the kinetics for adsorbate displacement of these molecules from an electrode surface as a function of incoming adsorbate concentration, structure, and size, all (26) (a) Caldwell, W. B.; Campbell, D. J.; Chen, K.; Herr, B. R.; Mirkin, C. A.; Malik, A.; Durbin, M. K.; Huang, K. G. J. Am. Chem. Soc. 1995, 117, 6071. (b) Herr, B. R.; Mirkin, C. A. J. Am. Chem. Soc. 1994, 116, 1157. (c) Shi, X.; Caldwell, W. B.; Chen, K.; Mirkin, C. A. J. Am. Chem. Soc. 1994, 116, 11598. (d) Mirkin C. A.; Caldwell, W. B. Tetrahedron 1996, 52, 5113-5130. (e) Caldwell, W. B.; Chen, K.; Mirkin, C. A.; Babinec, S. J. Langmuir 1993, 9, 1945. (f) Chen, K.; Caldwell, W. B.; Mirkin, C. A. J. Am. Chem. Soc. 1993, 115, 1193.

2172

Langmuir, Vol. 16, No. 5, 2000

Figure 1. Cyclic voltammetry of compounds 1 and 2 in 0.1 M n-Bu4NPF6/CH3CN. The scan rate is 200 mV/s.

Figure 2. Surface coverages of redox-active adsorbate 1 on a ceramic YBa2Cu3O7-δ electrode as a function of soaking time in solutions of 1. Solution concentrations are given in the inset.

of which can provide valuable insight into the nature of the exchange process (case 3 in Scheme 2). Growth of FcC(O)NH(CH2)4NH2 (1) Monolayer on Ceramic YBa2Cu3O7-δ. Monolayer growth was followed by immersing a ceramic YBa2Cu3O7-δ electrode (area ) 0.03 cm2) into a 1 mM CH3CN solution of 1 and determining surface coverage by cyclic voltammetry at various stages throughout the process. The coverage of 1, as determined by cyclic voltammetry, increased very steeply from 0 to 2 × 10-9 mol/cm2 during the first 2 h of immersion (Figure 2, 2). Then the growth process significantly slowed. After an additional 10 h of soaking, the coverage approached 2.8 × 10-9 mol/cm2, and after 21 h, it reached a value of 4.5 × 10-9 mol/cm2. After this experiment, the electrode was cut and sanded to generate a fresh surface (vide infra) for analogous kinetic studies in a 0.4 mM CH3CN solution of 1 (1). Kinetic studies in 4 mM (b) and 10 mM (9) CH3CN solutions of compound 1 were conducted in a similar manner on the same electrode. All data from these studies are presented in Figure 2. Generally speaking, the formation of a monolayer of 1 on a YBa2Cu3O7-δ surface, regardless of conditions, consists of two distinct periods of growth. The first few hours of soaking leads to a very fast increase in adsorbate surface coverage. Then the adsorption rate slows down, but a limiting surface coverage is never reached, regardless of concentrations of adsorbate and soaking time (up to 30 days), and the tailing part is described by a parabolic curve (coverage ∝ t1/2) (Figure 2). Over the range of adsorbate

Xu et al.

concentrations studied, final surface coverage values are directly proportional with soaking time and adsorbate concentration in the soaking solutions and time. From these studies, it is clear that the monolayer growth kinetics for compound 1 do not fit a simple Langmuir adsorption isotherm. For a Langmuir adsorption isotherm, the surface coverage traces are expected to level off as adsorbates saturate surface adsorption sites. On the basis of our studies, we propose that the adsorption of alkylamines on ceramic YBa2Cu3O7-δ is controlled by two major processes: (1) a relatively fast adsorption process, where the solution adsorbates adsorb onto external surface of the superconductor that is exposed to the solution, and (2) the diffusion of adsorbate molecules and solution through the electrode pores and channels, which are very characteristic of this type of ceramic material (Figure 3). Adsorbates then can adsorb onto the internal surface of the electrode with the effective electrode surface area increasing as electrolyte is wicked up into the material. Surface coverage values for 1 and differential capacitance measurements for an electrode that was modified by immersing it in a 1 mM solution of 1 are plotted as a function of reaction time (Figure 4). The differential capacitance, which was measured at -0.35 V vs Ag/AgNO3, decreased during the first ∼15 h of monolayer adsorption with a concomitant increase in surface coverage of 1. This is consistent with the expected passivation of the electrode surface as the monolayer of 1 forms. Importantly, the capacitance begins to increase thereafter, even though the surface coverage of 1 keeps increasing. This increase in differential capacitance with a simultaneous increase in surface coverage shows that the effective surface area is increasing as the electrode is exposed to solutioncontaining adsorbate. In addition to capacitance measurements, three other observations support this hypothesis. First, assuming that the cross sectional area of the ferrocenyl group (36 Å2) in 1 dictates its molecular footprint,12a a full monolayer of 1 on a perfectly flat substrate should be ∼4.5 × 10-10 mol/cm2. Surface coverages determined in our studies are on the order of 1 × 10-9 to 1 × 10-8 mol/cm2, depending on the adsorbate concentrations and soaking time. These coverages are 1-2 orders of magnitude higher than the theoretical value mentioned above, and normal roughness factors cannot account for these high values.27 However, ceramic superconductor YBa2Cu3O7-δ electrodes are quite porous, as evidenced by SEM (Figure 3). Therefore, we conclude that these anomalously high surface coverages and non-Langmuir isotherms are primarily due to adsorbate solution being wicked into the porous structure via capillary action. Adsorbate molecules can access surface sites on the interior of the ceramic electrode as this happens. Thus, the effective surface area includes both external and internal contributions. A second important observation pertains to experiments that involved long electrode soaking times. If electrodes treated in this manner were subsequently cut or sanded approximately 1 mm to generate a fresh surface, small waves attributed to 1 often could be detected by cyclic voltammetry. This is strong evidence that adsorbates do diffuse into the ceramic material and adsorb onto the internal surface of the porous electrode. Finally, the growth traces at different concentrations are also consistent with this hypothesis. In 0.4 mM solutions of 1, the first 2 h of soaking leads to an adsorbate surface coverage of 1.1 × 10-9 mol/cm2, which (27) (a) Trasatti, S.; Petrii, O. A. Pure Appl. Chem. 1991, 63, 711. (b) Alves, V. A.; da Silva, L. A.; de Castro, S. C.; Boodts, J. F. C. J. Chem. Soc., Faraday Trans. 1998, 94, 711.

Growth and Exchange Kinetics for Alkylamines

Langmuir, Vol. 16, No. 5, 2000 2173

Figure 4. Surface coverage of 1 (solid line, left abscissa) vs time, and differential capacitance (dashed line, right abscissa) vs time for a ceramic YBa2Cu3O7-δ electrode soaked in a 1 mM CH3CN solution of 1.

where D is the adsorbate diffusion constant, and ∂c/∂x is the concentration gradient. Note that because of this, at

higher concentrations (>10 mM), adsorbate diffusion through the pores of YBa2Cu3O7-δ and adsorption to internal sites likely contribute to a large portion of the electrochemically determined surface coverages, even for the first 2 h of soaking (Figure 2). Ten electrodes were studied to test the reproducibility of these results. Similar adsorption trends were observed in all studies. Although the surface coverages obtained for different electrodes modified under identical conditions differed by as much as 50%, surface coverage determination experiments involving one electrode, used repeatedly via the sanding and shaving method, exhibited reproducibility within 10%. This observation is undoubtly due to the structural complexity and porosity of the ceramic materials used to make the electrodes. Although surface Cu amine complexes with more than one amine ligand on each Cu atom might account for a surface coverage greater than the idealized case, it is unlikely to be responsible for the factor of 10 difference reported herein. If the system in question involved solution Cu amine complexes, this might be a plausible argument, but in the case of a surface, the exposed surface area, free volume, and the number of potential surface bonding sites dictate how much adsorbate can bond to the surface. Increasing the number of potential sites without increasing the available surface area and free volume could not account for a factor of 10 difference in adsorbate surface coverage. In fact, the number of copper sites is not the limiting factor in these systems. Molecular models show that the molecules are too large to occupy all available surface Cu sites.18b In other words, the Cu in this system is not coordinatively saturated with amine ligands and could accommodate other ligands, if there was room to do so. “Quasi-Degenerate” Exchange of Surface Adsorbed FcC(O)NH(CH2)4NH2 (1) with Solution Fc(CH2)6NH2 (2). This type of exchange is termed “quasidegenerate” since the chemistry of adsorption is degenerate in each case (primary alkylamine on YBa2Cu3O7-δ), and both alkylamine adsorbates (1 and 2) have ferrocenyl redox-active moieties of approximately the same size. With these molecules, the extent of exchange between 1 and 2 can be monitored easily through cyclic voltammetry. In a typical experiment, a ceramic YBa2Cu3O7-δ electrode (area ) 0.04 cm2) was soaked in a 1 mM CH3CN solution of 1 for 2 days.29 Cyclic voltammetry was

(28) Jost, W. Diffusion in Solids, Liquids, Gases; Academic: New York, 1960; Chapter 1, p 2.

(29) We chose 2 days of soaking to form the initial monolayer for experimental consistency.

Figure 3. SEM images of a typical ceramic YBa2Cu3O7-δ electrode: (A) as prepared; (B) after treatment with 600 grit sand paper, shaved with a razor blade, and sonicated in dry CH3CN; and (C) a pellet treated as described in B after soaking in a 1 mM CH3CN solution of 1 for 1 day.

is almost 80% of the coverage after 1 day of soaking (1.5 × 10-9 mol/cm2). In the case of 4 mM solutions of 1, 2 h of soaking generates only 45% of the surface coverage observed for 1 day of soaking (3.7 × 10-9 mol/cm2 vs 8.5 × 10-9 mol/cm2) (Figure 2). This concentration dependence is qualitatively consistent with the relationship between the adsorbate diffusion and concentration as predicted by Fick’s first law of diffusion,28

J ∝ -D ∂c/∂x

(2)

2174

Langmuir, Vol. 16, No. 5, 2000

Xu et al. Table 1. Wave Shape Analysis of the Quasi-Degenerate Exchange of 1 and 2 1

2

surface-confined 1 (%)

E1/2 (mV)a

∆Efwhm (mV)

100 (Figure 5A) ∼85 (Figure 5B) ∼70 (Figure 5C) ∼50 (Figure 5D) ∼30 (Figure 5E) ∼5 (Figure 5F)

320 320 320 340 320 NA

160 160 160 NA NA NA

E1/2 (mV) 80 75 90 85 90

∆Efwhm (mV)

E1/2 (mV)

NAb NA NA 150 170

240 245 250 235 NA

a All potential data reported in this tablew have an error of (10 mV. b No accurate data available due to the overlap of the cyclic voltammograms of 1 and 2.

Figure 5. Cyclic voltammograms of a YBa2Cu3O7-δ electrode modified with a monolayer of 1 after soaking in a 1mM CH3CN solution of 2. Scan rates are 200 mV/s and Y scales are 4 µm for all experiments.

performed on the modified electrode after it was removed from the soaking solution and rinsed with copious amounts of CH3CN. As expected, cyclic voltammetry performed on this electrode showed a near-ideal wave associated with the oxidation/reduction of surface adsorbed 1 at +320 mV vs Ag/AgNO3 (vide infra) (Figure 5A). This modified electrode was then immersed into a 1 mM CH3CN solution of adsorbate 2. The extent of exchange (2 for 1) was examined by cyclic voltammetry as a function of soaking time (Figure 5B-F). After 0.5 h the electrode was removed from the solution of 2, rinsed with CH3CN, and then placed into a clean electrolyte solution of 0.1 M n-Bu4NPF6/CH3CN. Cyclic voltammetry shows two reversible waves with E1/2 values at +80 and +320 mV vs Ag/AgNO3, which correspond to surface-confined 2 and 1, respectively. After each electrochemical measurement, the electrode was rinsed with CH3CN before it was immersed in a 1 mM CH3CN solution of 2. Within the first 0.5 h, approximately 15% of 1 was displaced by 2. After 10.5 h, there were nearly equal amounts of 1 and 2 on the electrode surface. This point is defined as the half-life of the exchange reaction. Significantly, near complete displacement (∼95%) of 1 by 2 took 10 days. Another important observation during the exchange process is that the sum of the current associated with surface-confined 1 and 2 increased as the electrode was exposed to the soaking solution. For example, after 10 days of the exchanging reaction (Figure 5F), the total integrated current associated with surface-confined 1 and 2 is 1.5 times that of the starting monolayer (Figure 5A). This is further evidence that the alkylamine (2) in solution, in addition to exchange with adsorbed 1, continues to diffuse through pores and channels to access new adsorption sites inside the ceramic material. The reversibility of the exchange process has been demonstrated in the following way. The electrode, now covered with a monolayer of 2 (with less than 5% of 1), was immersed into a 1 mM CH3CN solution of 1. After soaking for 1 day, it was removed from solution, rinsed,

and studied by cyclic voltammetry. The wave at +320 mV vs Ag/AgNO3 started growing at the expense of the one at +90 mV. Essentially this corresponds to the process involving the displacement of surface-confined 2 by 1, which, after more than 30 days, leads to a monolayer consisting of greater than 95% of 1. The cyclic voltammograms of the redox-active monolayers taken during the exchange reaction have been analyzed to probe the chemical environment of the adsorbates within the film. For an ideal surface-confined redox-active species, the peak current is linearly proportional to scan rate, the anodic wave is the mirror image of the cathodic wave reflected across the potential axis, and the full width at half-maximum of the peak (Efwhm) is 90.6 mV for a one-electron-transfer process.30 However, experimentally such ideal behavior is rarely observed due to a variety of factors, including local environment. Soaking a YBa2Cu3O7-δ electrode in a 1 mM CH3CN solution of 1 for 2 days results in a nonideal, but wellbehaved, voltammetric wave for a surface-confined species (∆Ep ) 50 mV, Efwhm,a ) 160 mV, ia/ic ) 0.98 at a scan rate of 200 mV/s) (Figure 5A and Table 1). For a mixed monolayer consisting of two adsorbates with different redox potentials, the anodic and cathodic peak positions of the adsorbate with higher formal potential may be affected by the change in oxidation state of the adsorbate with lower formal potential, depending upon proximity and degree of phase segregation.31 If 2 and 1 are in a densely packed monolayer, but mixed randomly at the molecular level on the surface, the oxidation of ferrocenyl to ferrocenylium [Fe(II) f Fe(III)] in 2 will influence nearby 1 due to electrostatic repulsions, which would make the oxidation of 1 more difficult (i.e. Ep,a of 1 shifts to more positive potentials, and the wave broadens).21,26,31 Peak shape analysis of the quasi-degenerate exchange reaction shows that, within experimental error, the formal potentials and relative peak positions (Ep,a values and ∆Ep,a) of the two adsorbates do not shift during the exchange process (Table 1). Furthermore, no significant peak broadening (Efwhm,a values) is observed during the 10 days of the exchange reaction. These data suggest that repulsive interactions between surface-confined oxidized 1 and 2 are not significant throughout the electrochemically assayed exchange process. Since these molecules are quite similar in structure, relatively poor adsorbate packing on these polycrystalline ceramic substrates, rather than phase segregation, is likely responsible for this response. “Nondegenerate” Exchange of a FcC(O)NH(CH2)4NH2 (1) Monolayer with Non-Redox-Active Alky(30) Murray, R. W. In Eletroanalytical Chemistry; Bard, A. J., Ed.; Marcel Dekker: New York, 1984; Vol. 13, pp 200-206. (31) (a) Tirado, J. D.; Acevedo, D.; Bretz, R. L.; Abruna, H. D. Langmuir 1994, 10, 1971. (b) Tirado, J. D.; Abruna, H. D. J. Phy. Chem. 1996, 100, 4556. (c) Acevedo, D.; Bretz, R. L.; Tirado, J. D.; Abruna, H. D. Langmuir 1994, 10, 1300.

Growth and Exchange Kinetics for Alkylamines

Figure 6. “Nondegenerate” exchange kinetics for surfaceconfined 1 and solution hexylamine. Surface coverages of redoxactive 1 are normalized to the coverages prior to the exchange reactions. Experiments were conducted on the same piece of ceramic electrode for two different hexylamine concentrations (inset).

lamines. Additional experiments were designed to gain mechanistic information regarding the exchange process. Unlike the aforementioned “quasi-degenerate” exchange reaction, which involves exchange of solution and surfacebound redox-labeled primary amines, this series of experiments involves the displacement of a surfaceconfined redox-active primary alkylamine 1 by redoxinactive primary, secondary, and tertiary alkylamine reagents in solution (case 3 in Scheme 2). These processes are termed “nondegenerate” because of the dissimilarity of the exchanging adsorbates. In a typical experiment, a redox-active monolayer of 1 was prepared on YBa2Cu3O7-δ, and the displacement of surface-confined 1 by redoxinactive alkylamine reagents (i.e. hexylamine) was monitored via cyclic voltammetry by measuring the decrease in current associated with the oxidation/reduction of 1 as a function of exposure to the redox-inactive alkylamines. This method allows one to monitor the signal exclusively due to the exchange reaction, since the diffusion of redoxinactive solution species into the pores during the exchange process will not show up in the electrochemical experiment. The mechanistic nature of this exchange reaction is likely quite complicated, but from such exchange studies, it may be possible to differentiate between associative and dissociative pathways. The latter has been reported to be the dominant mode of exchange for alkanethiol monolayers on Au22 and Pt.31 A diagnostic way of differentiating an associative process from a dissociative one is to monitor the reaction as a function of adsorbate concentration.32 The rate of an associative process will be dependent upon incoming adsorbate concentration, while a dissociative process will be relatively unaffected. A monolayer of 1 on a YBa2Cu3O7-δ electrode (Γ ) 3.8 × 10-9 mol/cm2) was prepared as described earlier. The electrode was removed, rinsed vigorously with CH3CN, and characterized by cyclic voltammetry before it was soaked in CH3CN solutions of hexylamine. Data at two different concentrations of hexylamine (1 and 10 mM) were collected. For comparison purposes, surface coverages of 1 as a function of soaking time were normalized to those found prior to initiating the exchange reactions (Figure 6). Significant differences between exchange rates were observed. For the 1 mM hexylamine solution, about 30% of surface-confined 1 was displaced after 0.5 h, while for (32) Steinfeld, J. J. Chemical Kinetics and Dynamics; Prentice Hall: Englewood Cliff, NJ, 1992; Chapter 1 and 2.

Langmuir, Vol. 16, No. 5, 2000 2175

Figure 7. “Nondegenerate” exchange data for surface-confined 1 with amines possessing different headgroups. Data on this figure and Figure 4 were collected from the same electrode.

a 10 mM solution, 45% was displaced after 0.5 h. The exchange rates slowed with longer soaking times. For example, after soaking the modified electrode in a 1 mM hexylamine solution for 20 h, 35% of the original monolayer 1 remained, while only 5% of monolayer 1 remained on the surface after soaking for 20 h in a 10 mM solution of hexylamine. The dependence of the exchange rates on the concentration of attacking hexylamine strongly indicates that the exchange reaction occurs through an associative pathway. To further test this hypothesis of an associative pathway, steric effects on the exchange dynamics were examined. We carried out nondegenerate exchange reactions of surface-confined 1 in 10 mM CH3CN solutions of N-methyl hexylamine and N,N′-dimethylhexylamine. The interactions of primary, secondary, and tertiary amines with YBa2Cu3O7-δ were previously shown to involve the same type of coordination chemistry.18,19 Substituting protons on the amino group of the attacking ligand with one or two methyl groups increases its size and might be expected to affect the rate of exchange. The exchange behavior of surface-confined 1 with secondary and tertiary amines is presented in Figure 7. During the first 0.5 h, 25% of 1 was displaced by N-methylhexylamine, while 40% of 1 was displaced by hexylamine at the same concentration. Actually, the displacement reaction curve for 1 mM hexylamine is very comparable to that observed for N-methylhexylamine at 10 mM concentration (Figures 6 and 7). The exchange rate dependence upon size of incoming molecule also is consistent with an associative mechanism because the rate of a reaction which proceeds by a dissociative mechanism is typically independent of the size of the incoming molecule. No significant rate differences were observed for reactions in solutions of N-methylhexylamine and N,N′-dimethylhexylamine, which may be an implication of a saturation of the steric effects. Electronically these molecules are quite similar, although not identical. When one compares the adsorbate exchange behavior reported herein for alkylamines and monolayer-modified YBa2Cu3O7-δ with the analogous data for alkanethiols on monolayer-modified Au, it appears somewhat surprising. The former is dominated by an associative pathway while the latter proceeds via a dissociative pathway.22 However, when one takes into account that the exchange chemistry on YBa2Cu3O7-δ is occurring at Cu(II) sites, it is actually somewhat predictable, especially when one examines the literature base revolving around the amine coordination chemistry of Cu(II). It is well-known that Cu(II) can accommodate coordination numbers ranging from 4 to 7

2176

Langmuir, Vol. 16, No. 5, 2000

and that the transitions among such coordination environments are often dynamic at room temperature.33,34 In fact, numerous solution studies have documented the susceptibility of Cu(II) to SN2 type displacement reactions35,36 like the type we are proposing to take place on the monolayer-modified reduced YBa2Cu3O7-δ surface with alkylamines. Conclusions We have carried out studies on the growth and exchange processes for alkylamine monolayers on the cuprate HTSC, YBa2Cu3O7-δ, and although the porous structure of this type of ceramic substrate significantly complicates the growth kinetics of redox-active alkylamine monolayers, several important conclusions may be drawn from these studies. First, there are two main factors that control the adsorption process and contribute to the adsorption (33) Gazo, J.; Bersuker, I. B.; Garaj, J.; Kabesova, M.; Kohout, J.; Langfelderova, J.; Melnik, M.; Serator, M.; Valach, F. Coord. Chem. Rev. 1976, 19, 253. (34) Basolo, F. B.; Pearson, R. G. Mechanisms of Inorganic Reactions; John Wiley and Sons: New York, 1967; pp 145-158, pp 421-422. (35) (a) Pearson, R. G.; Lanier, R. D. J. Am. Chem. Soc. 1964, 86, 765. (b) Pearson, R. G.; Anderson, M. M. Angew. Chem., Intnl. Ed. 1965, 4, 281. (36) (a) Pasternack, R. F.; Huber, P. R.; Huber, U. M.; Sigel, H. Inorg. Chem. 1972, 11, 276. (b) Sharma, V. S.; Leussing, D. L. Inorg. Chem. 1972, 11, 138.

Xu et al.

isotherms: (1) a rapid adsorption process that occurs on the external HTSC electrode surface and (2) a relatively slow pore diffusion process which results in the modification of the internal surface area of the substrate. Qualitatively, over millimolar concentration ranges, a major portion of alkylamine monolayer forms on the external surface of ceramic YBa2Cu3O7-δ within a few hours after which the pore diffusion process dominates, contributing to the parabolic adsorption isotherm. Second, alkylamine adsorption on YBa2Cu3O7-δ is based on Cu(II)-amine coordination chemistry and is chemically distinct from the previously studied systems involving thiol adsorption on Au and Pt. Accordingly, the exchange behavior of alkylamines on this novel substrate reflects much of the solution coordination chemistry of Cu(II).33-36 It is very dynamic and occurs via an associative rather than a dissociative pathway. Acknowledgment. C.A.M. acknowledges the AFOSR (F49620-96-1-0133) and the NSF (Grant DMR 91-20000) through the Science and Technology Center for Superconductivity for support of this work. C.A. M. also acknowledges use of instruments which were purchased by the MRSEC Program of the NSF at the MRC of NU, under Award No. DMR-9120521. Professor K. R. Poeppelmeier is thanked for use of instrumentation. LA990910Y